Although fetal development is one of the human growth periods most vulnerable to toxins, epidemiologic data concerning the effects of ambient air pollution exposures during this period are limited. Studies conducted previously in Los Angeles, China, Brazil, the Czech Republic, Mexico, and the United States suggest exposure to ambient air pollution during pregnancy may cause adverse birth outcomes, such as low birth weight (LBW) or intrauterine growth retardation (IUGR), preterm birth, and intrauterine and postneonatal mortality. We recently reported that air pollution, specifically carbon monoxide (CO) and particulate matter less than 10 microns in diameter (PM10), increased the risk of giving birth to LBW and preterm infants in the Los Angeles metropolitan area between 1989-1993 (Ritz and Yu, 1999). Further analyses also indicated increases in CO exposure during the second month of pregnancy may be related to certain cardiac defects. While our research and the Brazilian study suggest that motor vehicle-related pollution may be responsible for the observed effects, no study to date has examined this potential association. Furthermore, no information exists on whether exposures received in specific microenvironments, such as exposure from indoor sources and in-vehicle exposures while commuting are important. The objective of the proposed research is to determine whether exposure to elevated ambient air pollution during pregnancy results in LBW, preterm birth, intrauterine and postneonatal mortality, or cardiac defects in infants born to women living in the South Coast Air Basin (SoCAB). We will examine whether: (1) these effects are attributable to traffic-related sources relying on measures of proximity to high traffic density roadways; (2) short-term increases in air pollution prior to birth or fetal death and/or medium-term exposures accumulated over months or trimesters of pregnancy are more important; (3) observed effects are stable when adjusting for risk factor information not available in previous studies; (4) maternal exposures to additional indoor, in-vehicle, or occupational sources confound or modify the effect observed for ambient pollutants; and (5) effects are different for population subgroups (e.g., older mothers, working mothers etc.). We will use a two-tiered approach. First, we will perform a cohort study of all births (1994-1999) and fetal and infant deaths (1989-1997) in the SoCAB region and model short and long term exposures to pollutants on term LBW, preterm birth, intrauterine and postneonatal mortality and cardiac defects. Second, we will conduct a standardized mail survey for a nested group of 2000 LBW and/or preterm births to collect additional exposure, confounder, and effect modifier data for children born most recently. Finally we are collecting data that will allow us to validate the traffic density and ambient air pollution measures using a small sample of personal, indoor, and outdoor monitoring data for carbon monoxide and particulate matter.